To meet the continuing demand for high-capacity distributed data centers and the increasing need for scalable computing resources, the computer industry has pursued technologies to enable large-scale deployments of cloud computing environments. Cloud computing environments, which make use of virtualization technologies, such as virtual networking and virtual machine architectures, provide enhanced flexibility in the deployment of services and hardware infrastructures, by allowing for the abstraction of physical resources into logical representations. However, some of these attractive aspects of cloud computing environments can pose management challenges, for example, in the areas of provisioning, monitoring, and deploying such resources, because the very abstractions that provide flexibility can impede manageability.
With the arrival of cloud computing, network performance has become more dependent on efficient network management. Network management often determines the success or failure of a cloud offering.
Network management in a cloud environment can present additional challenges as compared to those in traditional IT management area. Facing dozens of or even hundreds of new customers on-boarding in a short time period and thousands of customers totally for overall management in the cloud, delivery of efficient and qualified network resource management in a way to provide rapid response, precise configuration and flexible control, based on various customer requirements and topology scenarios can be difficult to achieve.
Network administrators have been observed to perform manual-based network management for addressing customer network connection requests.
For the manual-based network approach, generally different network resource entities are configured by different internal network IT sub-teams a cloud service provider. This way is significantly time-consuming and usually error-prone, because there will be many communication efforts taken in the various inter-communication processes and precision will be potentially lost due to misunderstanding and a gap of communication in that process. Most important, it cannot adapt to the feature of the rapid provisioning of the cloud, especially under the scenario of several new customers requesting to be on-boarded in a short time.